emotional distress and prenatal attachment in pregnancy after perinatal loss

Journal of Nursing Scholarship Fourth Quarter 2002 339

Clinical Scholarship

Emotional Distress and Prenatal Attachment inPregnancy After Perinatal LossDeborah Smith Armstrong

Purpose: To evaluate the association of previous perinatal loss with parents’ levels of depressivesymptoms, pregnancy-specific anxiety, and prenatal attachment in a subsequent pregnancy,and to determine whether higher levels of depressive symptoms and pregnancy-specific anxietywere associated with prenatal attachment.

Design: A three-group comparative design was used to collect cross-sectional survey data. Thesample consisted of 103 couples who were in the second trimester of pregnancy: 40 coupleswho had a perinatal loss in a previous pregnancy, 33 couples were pregnant for the firsttime, and 30 couples had a history of prior successful pregnancies.

Methods: Structured questionnaires via in-person or telephone interviews were used to measuredepressive symptoms, pregnancy-specific anxiety, and prenatal attachment.

Findings: Couples with a history of perinatal loss had higher levels of depressive symptomsand pregnancy-specific anxiety than did couples with past successful pregnancies and nolosses; mothers had higher levels of symptoms than did fathers in all groups. Couples withand without a history of perinatal loss did not differ in their level of prenatal attachment inthe current pregnancy.

Conclusions: These findings do not support the theory that depressive symptoms and pregnancy-specific anxiety affect subsequent parent-infant attachment in a pregnancy after perinatalloss. However, they do provide insight into the continuing influence of parents’ previous lossexperience on their depressive symptoms and pregnancy-specific anxiety in subsequentpregnancies. Families should be assessed to examine the potential long-term influence ofemotional distress as a result of prior perinatal loss.

JOURNAL OF NURSING SCHOLARSHIP, 2002; 34:4, 339-345. ©2002 SIGMA THETA TAU INTERNATIONAL.

[Key words: perinatal loss, pregnancy, depressive symptoms, pregnancy-specific anxiety,prenatal attachment, emotional distress]

* * *

Perinatal loss is a traumatic event in the lives offamilies. It can have a major influence on thedevelopment of emotional distress, such as depressive

symptoms and pregnancy-specific anxiety, for both parents insubsequent pregnancies. Perinatal loss includes fetal death andneonatal death within the first 28 days of life. The incidence ofearly miscarriage before 20 weeks gestation ranges from 10 to20 per 100 pregnancies; late pregnancy loss is estimated at 2per 100 (Woods & Woods, 1997). In 1997, perinatal loss in theUnited States was reported as 11.6 per 1,000 live births (NationalCenter for Health Statistics, Vital Statistics for the United States,Centers for Disease Control and Prevention, 2001). The purposesof this study were to evaluate the association of previous perinatalloss with parents’ levels of depressive symptoms, pregnancy-specific anxiety, and prenatal attachment in a subsequentpregnancy, and to determine whether higher levels of depressivesymptoms and pregnancy-specific anxiety were associated withprenatal attachment.

Background

A substantial body of human and animal research indicatesthat affiliation and attachment between a mother and hernewborn affects the cognitive, affective, and behavioraldevelopment of the infant as well as other maternal andfamilial outcomes (Ainsworth, 1971; Bowlby, 1969; Hofer,

Deborah Smith Armstrong, RN, PhD, Iota Zeta, Assistant Professor,University of Louisville School of Nursing, Louisville, KY. This study wasfunded by grants from: Sigma Theta Tau International and American NursesFoundation, and a Woodrow Wilson/Johnson and Johnson DissertationFellowship. An additional small research grant was awarded from the IotaZeta chapter of Sigma Theta Tau International at the University of Louisville.The author gratefully acknowledges Dr. M.K. Raynes for her consultationon the data analysis. Correspondence to Dr. Armstrong, K-Building, Schoolof Nursing, University of Louisville, Louisville, KY 40292. E-mail:[email protected]

Accepted for publication March 18, 2002.

340 Fourth Quarter 2002 Journal of Nursing Scholarship

1994). In humans, affiliation and attachment between motherand child begins before birth (Cranley, 1981; Muller, 1993),and mothers’ prenatal attachment is positively correlated withattachment after birth (Bloom, 1995; Muller, 1996). Fathersalso develop prenatal attachment to their infants (Armstrong,2000; Ferketich & Mercer, 1995; Weaver & Cranley, 1983).Although mothers experience a more physical relationshipwith the unborn baby, fathers’ prenatal attachment is moreabstract (Sandelowski & Black, 1994).

Several factors influence the nature and development ofattachment during the prenatal period. Research on the effectsof psychosocial variables such as anxiety and depression onparents’ prenatal attachment has yielded inconsistent findings(Armstrong & Hutti, 1998; Condon & Corkindale, 1997).Further study to clarify these relationships is neededparticularly related to history of previous perinatal loss.

Research on the influence of perinatal loss has indicateddifficulties with emotional adjustment. The meaning of thepregnancy and the experience of perinatal loss influenceparents’ response to the loss and their grieving. In a qualitativestudy of 12 mothers and fathers who had experiencedmiscarriage, Hutti (1992) explored the meaning of thepregnancy and the loss. Three factors that influenced theintensity of the grief response were identified: (a) the realityof the pregnancy and the baby within, (b) congruence betweenthe actual loss experience and the standard of the desirable(“How it ought to be if I have to go through it”), and (c) theability of parents to make decisions to increase this congruence.A strong grief response was observed in parents for whom thepregnancy and the baby were perceived as real, whose actualexperience of loss was very different from the way they wishedit to be, and who saw themselves as unable to do anything tochange the situation.

Few studies have evaluated the traumatic effects of miscarriageon expectant fathers. With a convenience sample of 126expectant fathers whose partners experienced miscarriage before25 weeks gestation, Johnson and Puddifoot (1996) evaluatedlevel of grieving and effects of the miscarriage. They found ahigh level of grief in their sample of expectant fathers, similarto that reported for women (Toedter, Lasker, & Alhadeff, 1988).In addition, considerable psychological effect of the loss wasfound for a majority of men in the survey. These findings supportHutti’s (1992) model linking the meaning of the previouspregnancy and the emotional investment in the baby with theresponse to the loss.

Although the influence of previous perinatal loss duringsubsequent pregnancies is gaining recognition (Armstrong &Hutti, 1998; Cote-Arsenault & Mahlangu, 1999; Franche &Mikail, 1999; Theut, Pederson, Zaslow, & Rabinovich, 1988),few studies include both parents. Franche and Mikail (1999)examined the effects of previous loss on subsequent pregnancyby comparing expectant mothers and fathers with a historyof loss in their second trimester with expectant parents withno history of previous loss. Those with prior losses hadsignificantly higher levels of depressive symptoms and anxietythan did those without a history of loss. However, this studydid not address the influence of increased emotional distress

in pregnancies after perinatal loss on birth outcomes or onthe developing relationship between parents and their unborninfant.

Armstrong (2001) explored the meaning of a pregnancyafter previous loss in a qualitative study of four fathers duringpregnancy subsequent to that loss. Fathers described the effectsof the previous loss as well as the influence of their emotionaldistress on their experiences during the current pregnancy.Regardless of the timing of the loss or emotional investmentin the previous baby, all fathers in this small sample expressedgreater anxiety than in any previous pregnancy. Theydescribed a heightened sense of risk and a greater vigilancethan in prior pregnancies.

Prior research describes the relationship between themeaning of the previous pregnancy and parents’ response tothe perinatal loss (Armstrong, 2001; Hutti, 1992). This responsemay contribute to the development of depressive symptomsand anxiety about the outcome of a subsequent pregnancy(Armstrong & Hutti, 1998; Franche & Mikail, 1999).Childbirth is a significant, joyful experience for most families,but those who have suffered prior perinatal losses may viewsubsequent pregnancies with fear and a general hesitancy toanticipate positive outcomes (Armstrong, 2001; Cote-Arsenault& Mahlangu, 1999). Clinical studies have been focused ongrief responses after perinatal loss (Hutti, 1992; Swanson-Kaufman, 1986) and interventions to facilitate adaptation forparents at the time of the loss (Brost & Kenney, 1992; Swanson,1999). However, little attention has been given to parents’experience during a subsequent pregnancy or on their concernsabout the outcome of the pregnancy and the effect of emotionaldistress on prenatal attachment. If higher levels of depressivesymptoms and anxiety about the outcome of the currentpregnancy diminish prenatal attachment, evaluation of arelationship between this emotional distress and prenatalattachment is important for health care providers in theirevaluation of potential risks for the development of a positiveparent-child relationship.

Methods

Design and SampleA three-group comparative design was used to collect cross-

sectional survey data via in-person and telephone interviews.The sample was 103 couples in the second trimester of themother’s pregnancy. Forty couples that had previous perinatallosses and were pregnant again were compared with 33couples pregnant for the first time and with 30 couples with ahistory of prior successful pregnancies.

The volunteer sample was recruited from prenatal clinicsand education classes and private medical practices. Perinatal-loss support groups and newsletters nationwide also wererecruitment vehicles. In addition, notices about the study wereposted on Internet message boards focused on perinatal lossand general pregnancy, and interested parents were recruitedfrom those responding. Fifty-one percent of parentsexperiencing their first pregnancy and 63% of those with past

Emotional Distress


successful pregnancies were residents of Kentucky. However,only 13% of the participants with a history of perinatal losscurrently resided in Kentucky.

Both mothers and fathers were recruited for this study.Participants were 18 years of age or older and read andunderstood English. For the loss group, previous loss at anygestational age was accepted. Parents in the loss group wereexcluded from the study if they had experienced successfuldelivery of a healthy infant in a pregnancy subsequent to theloss. Parents were in a coupled relationship, as they defined it.

At an alpha level of .05 and with a minimum of 30 couplesin each group, the analysis of variance (ANOVA) F test hadapproximately 55% power to detect a medium effect and93% power to detect a large effect (Cohen, 1988). At asignificance level of .05 and with a minimum sample size ofat least 30 couples, the paired t test had approximately 55%power to detect a significant difference between fathers andmothers, assuming a medium effect and a correlation betweenmale and female scores as small as .40 (Cohen, 1988).

The parents (N=206) were mostly Caucasian (90%),married (93%), and had upper-middle income. Most werecollege educated. Participants ranged in age from 18 to 45years. Parents in the loss group were slightly older (M=32.7years, SD=5.5) than parents in either of the two nonloss groups(first pregnancy M=29.7, SD=5.6; past pregnancy and nolosses M=29.5, SD=5.0; F2,205=7.66, p=.001). The three groupsdid not differ by race, marital status, or educational level.Parents with previous successful pregnancies and no losseshad a lower annual household income than did either of theother two groups of parents (χ2[6, N=206]=12.8, p<.05).Parents in the loss group had experienced an average of twoprior perinatal losses (range=1 to 7). The mean gestationalage at the time of the loss was 22.6 weeks (SD=12.3). Twenty-nine percent of the losses occurred in the first trimester, 35%occurred in the second trimester, and the remaining 36% werelate pregnancy losses. At the time of the study interview, themean gestational age in all groups was 24.5 weeks (SD=5.1).

Measures

Depressive symptoms. The Center for Epidemiologic Studies-Depression Scale (CES-D; Radloff, 1977) is a 20-item self-reportscale developed to identify the duration and frequency ofdepressive symptoms. Respondents rated each symptomexperienced during the previous week on a 4-point scale rangingfrom 0 (rarely) to 3 (all the time). Scores of 16 or greater indicatea high level of depressive symptoms. This score correspondedto the 80th percentile of scores in community samples (Comstock& Helsing, 1976). The CES-D has shown high internalconsistency reliability with Cronbach’s alphas of .84 to .91 andgood test-retest reliability (Comstock & Helsing, 1976; Hall,Kotch, Browne, & Rayens, 1996; Radloff, 1977). This scalehas been used frequently to identify the presence of depressivesymptoms in pregnant and parenting women (Hall et al., 1996;Logsdon & Usui, 2001; Sachs, Hall, Lutenbacher, & Rayens,1999). Cronbach’s alphas in this study were .91 for the mothers

and .85 for the fathers, indicating acceptable internal consistencyof this scale in this sample.

Pregnancy-specific anxiety. The Pregnancy OutcomeQuestionnaire (POQ; Theut et al., 1988) is a measure ofanxiety about the outcome of the current pregnancy. The 15-item POQ indicates concerns related to the course ofpregnancy. The items on the questionnaire were derived frominterviews with expectant parents with and without historyof perinatal loss. Responses are scored on a 4-point Likertscale with response options ranging from 1 (almost never) to4 (almost always). Internal consistency reliability has beenreported with alphas ranging from .80 to .89 (Armstrong &Hutti, 1998; Theut et al., 1988). Cronbach’s alphas for thisstudy were .88 for mothers and .77 for fathers.

Prenatal attachment. The Prenatal Attachment Inventory (PAI;Muller, 1993) was designed to measure the emotional affiliationdeveloping between mother and child before birth. The 21-itemquestionnaire has no subscales and is scored on a 4-point Likertscale with response options of 1 (almost never) to 4 (almostalways). Internal consistency of the PAI was demonstrated inprior research with alphas ranging from .81 to .93 (Armstrong& Hutti, 1998; Muller, 1993; Muller, 1996). In this study,Cronbach’s alpha for the mothers was .87.

The PAI was modified for fathers (PAI-F; Armstrong, 2000).Items were reworded to indicate gender differences. The PAI-F has 22 items and is scored on a Likert scale similar to thePAI for mothers. Expectant fathers and a panel of expertswho worked with these fathers evaluated the items of the PAI-F for content validity. Prior support for the internal consistencyof the modified PAI-F was demonstrated (alpha of .87;Armstrong, 2000). In the current study, the Cronbach’s alphafor the fathers was .82.

Procedures

The study was approved by the university’s institutionalreview board. Prospective participants were identified by theirprimary health care provider or by self-selecting in responseto recruitment messages or flyers. For uniformity of datacollection related to the length of gestation of the currentpregnancy, interviews were completed during the 16th to 32ndweek of pregnancy (M=24.5, SD=5.1). Data collected fromone member of the couples were paired with the partnersusing predetermined identification numbers.

Data Analysis Three separate one-way analysis of variance (ANOVA)

models were used to determine whether depressive symptoms,pregnancy-specific anxiety, and prenatal attachment differedamong the three groups. Three separate two-way repeatedmeasures ANOVAs were used to evaluate differences indepressive symptoms, pregnancy-specific anxiety, and prenatalattachment by gender of the parent and group membership.A nested, repeated measures design was used for multivariateanalysis to account for any error variance that might be relatedto the nonindependence of the study participants’ responses

Emotional Distress


because of the potential correlation between scores of the twoparents in each couple.

Findings

The mean CES-D score for all parents combined was 11.2(SD=9.4). Among all groups, a greater proportion of themothers (38%) had CES-D scores of 16 or above, comparedto 17% of the fathers (�2 [1, N=206]=1.87, p<.0001). Forparents in the loss group, 45% of the mothers but only 23%of the fathers had CES-D scores of 16 or above (�2 [1,N=80]=4.53, p< .05). A large percentage of first-time mothers(30%) compared to first-time fathers (9%) had depressivesymptoms in the high range (�2 [1, N=66]=4.69, p<.05).

Posthoc analyses for pairwise comparisons of groupdifferences in depressive symptoms revealed that parents withprior perinatal losses had higher depressive symptoms thandid parents with no previous pregnancies. No significantdifferences in depressive symptoms were found among theother groups (see Table 1).

Pregnancy-specific anxiety also differed by group. Parentswith previous losses had significantly higher prenatal anxietyabout the outcome of the current pregnancy than did expectantparents with no history of perinatal loss. The two nonlossgroups did not differ in level of pregnancy-specific anxiety.No significant group differences in prenatal attachment werefound (see Table 1).

In the comparison of all mothers with all fathers, mothershad higher mean levels of depressive symptoms, anxiety aboutthe outcome of the current pregnancy, and prenatal attachmentthan did fathers. Significant differences in depressive symptomswere found between mothers and fathers in the loss group,

Emotional Distress

Table 1. Mean Differences in Depressive Symptoms,Pregnancy-Specific Anxiety, and Prenatal Attachment byGroup (N=206)

SignificantVariable Groupsa Mean (SD) F2, 205 Scheffé

comparisonsDepressive symptoms

1. Previous loss 13.34 1(9.5) 3.63** 1, 2*2. First pregnancy 19.32 1(8.3)3. Past pregnancies 10.50 (10.0)

no losses

Pregnancy-specific anxiety1. Previous loss 32.28 1(7.9) 21.90*** 1, 2***2. First pregnancy 26.17 1(5.9)3. Past pregnancies 25.22 1(6.8) 1, 3***

no losses

Prenatal attachment1. Previous loss 54.19 (10.9) 1.96**2. First pregnancy 57.12 1(9.1)3. Past pregnancies 56.88 1(9.4)

no lossesa Previous loss group n=80; First-pregnancy group n=66; Past pregnancies/no losses n=60.*p <.05; ***p<.0001.

but not for parents with previous pregnancies and no historyof loss. Significant gender differences in pregnancy-specificanxiety were found between parents with a history of perinatalloss. Mothers reported greater anxiety about the outcome ofthe current pregnancy than did fathers in this group. Differencesbetween mothers and fathers in the two nonloss groups werenonsignificant. For prenatal attachment, the only significantdifference was between mothers and fathers with pastsuccessful pregnancies without perinatal losses. In this group,mothers reported higher prenatal attachment than did fathers(see Table 2).

Table 2. Mean Differences in Depressive Symptoms, Pregnancy-Specific Anxiety, and Prenatal Attachment byGender and Group (N=206)

Mothers (n=103) Fathers (n=103)Variable/group Mean SD Mean SD t df

Depressive symptomsAll parents 14.06 10.4 18.39 7.4 -4.52** 204

Previous loss 16.48 19.7 10.20 8.3 -3.10** 178First-pregnancy 12.42 19.4 16.21 5.6 -3.26** 164Past pregnancy/no loss 12.63 11.9 18.37 7.4 -1.67** 158

Pregnancy-specific anxietyAll parents 29.75 18.6 26.78 6.2 -2.83** 204

Previous loss 34.98 18.5 29.56 6.4 -3.23** 178First-pregnancy 27.10 16.2 25.24 5.3 -1.30** 164Past pregnancy/no loss 25.70 17.8 24.73 5.8 -.54** 158

Prenatal attachment All parents 58.32 10.1 53.50 9.3 -3.56** 204 Previous loss 56.53 12.2 51.85 9.1 -1.94** 178 First-pregnancy 58.76 19.1 55.45 8.9 -1.49** 164 Past pregnancy/no loss 60.23 17.7 53.53 9.8 -2.49** 158

aPrevious loss group n=80; First-pregnancy group n=66; Past pregnancies/no losses n=60.*p <.05; **p<.001.


No significant correlations were found between gestationalage at the time of the interview and level of depressivesymptoms for any group or gender. Gestational age at thetime of the previous loss was not significantly associated withlevel of depressive symptoms for parents with previous losses.For parents with a history of perinatal loss, as gestationalage at the time of the prior loss increased, their anxiety aboutthe outcome of the current pregnancy also increased (r=.23,p<.04). In addition, as gestational age in the current pregnancyincreased at the time of the study interview, anxiety aboutthe outcome of the current pregnancy decreased (r=-.26, p=.02).No significant association was found between pregnancy-specific anxiety and gestational age at the time of the interviewfor either group of parents without a history of previous loss.As expected, for all parents, as gestational age at the time ofthe interview increased, their prenatal attachment increased(r=.32, p=.004).

The three separate two-way repeated measures ANOVAsto evaluate differences in depressive symptoms, pregnancy-specific anxiety, and prenatal attachment by gender of theparent and group membership showed higher depressivesymptoms and pregnancy-specific anxiety among parents inthe loss group than among parents in either of the two nonlossgroups. The mothers had higher mean levels of depressivesymptoms and pregnancy-specific anxiety compared to thecombined group of fathers. Mothers also had a higher meanlevel of prenatal attachment than did fathers. No significantdifferences were found in depressive symptoms or pregnancy-specific anxiety between the two nonloss groups. Prenatalattachment did not differ by group membership.

Discussion

Parents with a history of perinatal loss had a higher levelof depressive symptoms than did parents experiencing theirfirst pregnancy. In addition, among all parents, mothersreported greater depressive symptoms than did fathers.Although the mean CES-D score for expectant parents with ahistory of perinatal loss was below the score indicatingevidence of depressive symptoms, 45% of the mothers and23% of the fathers had depressive symptom scores of 16 orabove. This finding is consistent with the report of Francheand Mikail (1999) that mothers and fathers with a history ofperinatal loss had greater depressive symptoms in subsequentpregnancies compared with nonloss parents.

Gender and group differences also were found in levels ofpregnancy-specific anxiety. Armstrong and Hutti (1998) andCote-Arsenault and Mahlangu (1999) also found higherpregnancy-specific anxiety for expectant mothers in pregnanciessubsequent to perinatal loss. These findings also coincide withthose of Franche and Mikail (1999) who reported increasedprenatal anxiety for both expectant mothers and fathers in apregnancy after loss. However, the current findings are differentfrom those of Theut and colleagues (1988) who found nodifference in levels of pregnancy-specific anxiety for expectantfathers with prior losses compared to those experiencing their

first pregnancy. These conflicting findings could be a result ofdifferences in timing of the data collection. Theut and colleagues(1988) evaluated parents in the 8th month of pregnancy whendifferences in anxiety about the outcome of the current pregnancymay have decreased to a level not significantly different fromthose without previous losses.

No differences were found among the groups on prenatalattachment. Mothers had higher levels of prenatal attachmentcompared with fathers across all groups. This finding maybe a result of the physical relationship between mothers andtheir unborn infants during pregnancy. In addition, themodified prenatal attachment questionnaire for fathers mightbe inadequate to measure the full extent of fathers’ developingattachment with the unborn baby. These findings may bereassuring for parents with prior losses because, despiteincreased psychological distress, this sample indicated limitedinfluence of this distress on their developing attachment withtheir unborn infants.

Several of the study’s limitations warrant consideration.The volunteer sample might not be representative of expectantparents at large, diminishing the generalizability of findings.Parents with previous losses who volunteered to participatein the study are not representative of all parents who have ahistory of pregnancy loss. Therefore, the findings might reflectonly the experiences of those parents who are most affectedby their loss or are interested in sharing their stories. Thesample was fairly homogenous with respect to race, incomelevel, and marital status, therefore limiting the ability toevaluate differences between parents of other racial, ethnic,and socioeconomic backgrounds. Parents in the loss groupreported a relatively high level of annual household income,so evaluation of distress among parents with lower incomelevels was not possible. In addition, because only couples incommitted partnerships were recruited, the effects of a previousperinatal loss or the influence of emotional distress duringpregnancy on the developing prenatal attachment for parentswho were not in such a relationship could not be evaluated.The cross-sectional design of this study did not allow strongcausal inference among the variables, evaluation of changinglevels of pregnancy-specific anxiety or depressive symptomsduring the course of the pregnancy, or assessment of parents’developing prenatal attachment over time.

No relationship was found between the psychologicaldistress of parents in pregnancy after perinatal loss and theirdeveloping relationship with their unborn infant. However,prior research has indicated efforts by some parents to delayattachment to the current infant (Armstrong, 2001; Armstrong& Hutti, 1998; Cote-Arsenault & Mahlangu, 1999).Evaluating early development of parent-infant relationshipsafter birth is important to identify what, if any, effects previouspregnancy loss or other traumatic experiences in priorpregnancies may have on future parent-infant attachment.Heller and Zeanah (1999) investigated prior perinatal lossand disordered attachment with subsequent infants andreported 45% of these children had disturbed attachmentrelationships with their mothers at 12 months of age.Replicating studies such as this can indicate the association

Emotional Distress


between past traumatic experiences for mothers and fathersand early parent-infant attachment relationships.

Conclusions

Prior perinatal loss can be traumatic during subsequentpregnancies for expectant parents. This study adds to the growingbody of knowledge about parents’ experience of pregnancy afterprior perinatal loss. These findings challenge the belief thatonce a new pregnancy is achieved, concerns and grievingdecrease. In contrast, the emotional distress of these parentswho had experienced a prior loss was significant.

Methods to reduce psychological distress should be testedin future research. The development and evaluation ofinterventions to decrease anxiety is important for families asthey go through the bittersweet experience of subsequentpregnancies. The current findings support the position that aneed exists for a better understanding of ways to addressparents’ concerns and fears in pregnancies after perinatal loss.

For healthcare practitioners working with expectant families,the findings show the importance of evaluating prior obstetrichistory to examine the influence of prior perinatal loss onparents’ anxiety and depressive symptoms in subsequentpregnancies. Support and referrals to mental health servicesmay be needed. The results of this research support pastfindings highlighting the complexity of pregnancy afterperinatal loss and indicate the long-term effects of such losses.Practitioners need to create a caring environment in whichparents feel free to discuss their fears, validate their losses,and separate their past experience from the current pregnancy.

Healthcare providers should support and encourage anactive role by expectant fathers during pregnancy, and shouldrecognize differences between mothers and fathers in grievingthe previous loss and the experience of subsequent pregnancy.Although mothers’ experiences of the loss and subsequentpregnancy are more physical and less abstract, fathers strugglewith their multiple roles. The current findings as well asprevious research support the position that prior pregnancyloss influenced expectant fathers. While they grieve the lossof a wished-for child, fathers also may feel powerless tosupport and protect their families. Hypervigilance insubsequent pregnancies and heightened involvement may beways fathers can attempt to maintain some control. Providinginformation for fathers and involvement during the course ofthe pregnancy can address fathers’ concerns as well as theirability to be supportive to their partners.

Understanding the emotional ups and downs that parentsexperience during pregnancy after perinatal loss can provideinsight into the needs of these families at this critical time.The findings of this research should heighten health careprofessionals’ awareness and deepen their understanding ofthe mixture of hope and fear expectant parents experienceduring pregnancies subsequent to perinatal loss. Addressingthis emotional distress may influence the course of the currentpregnancy as well as future parent-infant relationships forthese developing families.

References

Ainsworth, M.D. (1971). The development of infant-motherattachment. In B.M. Caldwell & H.N. Ricciuti (Eds.). Review ofchild development research (pp.1-94). Chicago: University ofChicago Press.

Armstrong, D. (2000). Modification of the prenatal attachmentinventory for use with fathers. Unpublished manuscript, Universityof Kentucky, Lexington, KY.

Armstrong, D. (2001) Exploring father’s experience of pregnancyafter a prior perinatal loss. MCN, The American Journal ofMaternal Child Nursing, 26, 147-153.

Armstrong, D., & Hutti, M. (1998). Pregnancy after perinatal loss:The relationship between anxiety and prenatal attachment in asubsequent pregnancy. Journal of Obstetric, Gynecologic, andNeonatal Nursing, 27, 183-189.

Bloom, K. (1995). The development of attachment behaviors inpregnant adolescents. Nursing Research, 44, 284-289.

Bowlby, J. (1969). Attachment and loss: Attachment (Vol. 1). NewYork: Basic Books.

Brost, L., & Kenney, J. (1992). Pregnancy after perinatal loss: Parentalreactions and nursing interventions. Journal of Obstetric,Gynecologic, and Neonatal Nursing, 21, 457-463.

Cohen, J. (1988). Statistical power analysis for the behavioral science.Mahwah, NJ: Lawrence Erlbaum.

Comstock, G.W., & Helsing, K.T. (1976). Symptoms of depressionin two communities. Psychological Medicine, 6, 551-563.

Condon, J.T., & Corkindale, C. (1997). The correlates of antenatalattachment in pregnant women. British Journal of MedicalPsychology, 70, 359-372.

Cote-Arsenault, D., & Mahlangu, N. (1999). Impact of perinatalloss on the subsequent pregnancy and self: Women’s experience.Journal of Obstetric, Gynecologic, and Neonatal Nursing, 28, 274-282.

Cranley, M. (1981). Development of a tool for the measurement ofmaternal attachment during pregnancy. Nursing Research, 30,281-284.

Ferketich, S., & Mercer, R. (1995). Paternal-infant attachment ofexperience and inexperienced fathers during infancy. NursingResearch, 44, 31-37.

Franche, R., & Mikail, S. (1999). The impact of perinatal loss onadjustment to subsequent pregnancy. Social Science and Medicine,48, 1613-1623.

Hall, L.A., Kotch, J.B., Browne, D., & Rayens, M.K. (1996). Self-esteemas a mediator of the effects of stressors and social resources on depressivesymptoms in postpartum mothers. Nursing Research, 45, 231,238.

Heller, S.S., & Zeanah, C.H. (1999). Attachment disturbances in infantsborn subsequent to perinatal loss: A pilot study. Infant Mental HealthJournal, 20, 188-199.

Hofer, M.A. (1994). Early relationships as regulators of infant physiologyand behavior. Acta Paediatrica, (Supp. 397), 9-18,

Hutti, M. (1992). Parents’ perceptions of the miscarriage experience. DeathStudies, 16, 401-413.

Johnson, M.P., & Puddifoot, J.E. (1996). The grief response in partners ofwomen who miscarry. British Journal of Medical Psychology, 69, 313-327.

Logsdon, M.C., & Usui, W. (2001). Psychosocial predictors of postpartumdepression in a diverse group of women. Western Journal of NursingResearch, 23(6), 563-574.

Muller, M. (1993). Development of the prenatal attachment inventory.Western Journal of Nursing Research, 15, 199-215.

Muller, M.E. (1996). Prenatal and postnatal attachment: A modestcorrelation. Journal of Obstetric, Gynecologic, and Neonatal Nursing,25, 161-166.

National Center for Health Statistics. (2001). National vital statisticssystem, fetal deaths. Retrieved August 2002, from www.cdc.gov/nchs/nvss.htm

Radloff, L. (1977). The CES-D Scale: A self-report depression scale forresearch in the general population. Applied Psychological Measurement,1, 385-401.

Emotional Distress


Sachs, B., Hall, L.A., Lutenbacher, M., & Rayens, M.K. (1999).Potential for abusive parenting by rural mothers with low-birth-weight children. Image: Journal of Nursing Scholarship, 31, 21-25.

Sandelowski, M., & Black, B.P. (1994). The epistemology of expectantparenthood. Western Journal of Nursing Research, 16, 601-614.

Swanson, K.M. (1999). Effects of caring, measurement, and time onmiscarriage impact and women’s well being in the first year subsequentto the loss. Nursing Research, 48, 288-298.

Swanson-Kaufman, K.M. (1986). Caring in the instance of unexpectedearly pregnancy loss. Topics in Clinical Nursing, 8, 37-46.

Theut, S., Pederson, F., Zaslow, M., & Rabinovich, B. (1988). Pregnancysubsequent to perinatal loss: Prenatal anxiety and depression. Journalof the American Academy of Child and Adolescent Psychiatry, 27,289-292.

Toedter, L., Lasker, J., & Alhadeff, J. (1988). The Perinatal Grief Scale:Development and initial validation. American Journal ofOrthopsychiatry, 58, 435-449.

Weaver, R.H., & Cranley, M.S. (1983). An exploration of paternal-fetalattachment behavior. Nursing Research, 32, 68-72.

Woods, J., & Woods, J. (1997). Loss during pregnancy or in the newbornperiod. Pitman, NJ: Jannetti.

Emotional Distress

emotional distress and prenatal attachment in pregnancy after perinatal loss

Documents